Benefits, Rights, and Features · benefits available under the plan. Rights and features are all other rights and features available to any employee under the plan. So it's not the

RECORD OF SOCIETY OF ACTUARIES 1995 VOL. 21 NO. 3A

BENEFITS, RIGHTS, AND FEATURES

Moderator: Panelist:

PETER R. STURDIVAN PETER R. STURDIVAN

A plan design seminar session--are your plans discriminatory? Thought you were in a safe harbor? Like an iceberg, 90% of discrimination lies hidden beneath the surface. Learn about a "forgotten "feature of the 401(a)(4) regulations.

MR. PETER R. STURDIVAN: I work with Milliman & Robertson in Portland, OR. My area of emphasis is defined-benefit (DB) plans, specifically multiemployer plans, public plans, and plans sponsored by tax-exempt corporations.

This session will have a lecture format, and then I 'd like to see if we can have discussion on the issues that have come up since we have been through the determination letter process. I am going to provide you information so that you will be able to take the plans that you consult on and apply this general rule: benefits, rights, and features (BRFs) are made available in a nondiscriminatory manner only if each BKF is currently and effectively available.

There are four objectives in applying that general rule. First is to define what BRFs are. As actuaries, that's been kind of a problem, because we do tend to focus on the nondiscriminatory nature of benefit amounts and plan amendments. Section 401(a)(4)-4 gets overlooked to a certain extent, as the booklet described. It 's forgotten about, but it is a part of the nondiscrimination regulations. Therefore, it's important because failure can be a qualification issue. Second, I will define and test current availability. As you know, if you have worked through the determination letter filing, we actually can test current availability, and the IRS will review it. Third, I will define effective availability. The IRS will not review effective availability on a determination letter filing, but on subsequent audit there could be potential compliance issues. Finally, there are some special rules and we're going to see how those are applied.

First let's define what BKFs are. Benefits are all optional forms of benefits and ancillary benefits available under the plan. Rights and features are all other rights and features available to any employee under the plan. So it 's not the amount of benefit, but the form of benefit or ancillary benefit.

All optional forms of benefits are §411 (D)(6)(A) forms of benefit, including the normal form of benefit, and distribution alternatives that are early-retirement benefits or a §41 l(d)(6)(B)(I) retirement-type subsidy, including a qualified Social Security supplement. This is a revisit of § 1.411 (d)(6). There are 11 examples of what the IKS considers a form of benefit. If you want to get that out and take a look at it, that's a good review.

A distribution alternative, that is, an early-retirement benefit, such as a §411 (d)(6)(B)(I) retirement-type subsidy, appears to include, as an example, subsidized early-retirement benefits.

When do you know you have different optional forms of benefit? The regulation lists relevant terms that we can apply in looking and reviewing a plan document or a summary

373

RECORD, VOLUME 21

plan description (SPD) to determine whether there are different optional forms of benefit in that plan.

Relevant terms include the method of the benefit calculation, the actuarial assumptions used to determine an optional form of benefit (for example, the underlying assumptions for the joint survivor factors), payment schedules (for example, a different loan repayment schedule for the highly compensated than for the non-highly-compensated), timing, benefit commencement (that would be a relevant term in deciding whether there are two different optional forms of benefit), and medium &distribution. (Is the distribution in cash or in kind?)

Other relevant terms include election rights, difference in eligibility rights (for example, early-retirement eligibility can be different for different divisions), portions of a benefit in which the different alternatives apply, and different normal retirement ages. I 'm not referring to the uniform normal retirement age of the later of 65 and the fifth anniversary of participation. For example, what i ra plan had a normal retirement age based on the Social Security retirement age? At 65, 66, and 67? That might be considered a different normal retirement age. You may have to test that. Differences in the normal form of benefit is another relevant term. These are what the IRS deems to be relevant terms when you're reviewing a plan and trying to determine whether the plan has different optional forms of benefit.

Almost equally and almost more important, in my opinion, are the exceptions that are available when you're trying to determine whether the plan has different optional forms. Differences in benefit formulas, accrual method, or allocation formula are exceptions when you're reviewing optional forms of benefits. This becomes very important in reviewing an early-retirement window. Service computation periods can be different, definitions of compensation can be different. Vesting schedule differences can be ignored when you're determining different optional forms of benefit. Other exceptions include differences in allocation of gains and losses in DC plans, a plan that uses a different interest rate for lump sums that are valued in excess of $25,000, and finally, again, I alluded to this earlier, differences attributable to the uniform normal retirement age.

The regulations give a couple examples. Qualified joint and survivor factors are different for employees of one division as compared with employees of another division. That constitutes, even though it 's the same optional form, two different optional forms of benefit, and you're going to have to test it. By the same token, a plan that has two different benefit formulas but offers lump sums on the same assumptions and terms to both groups of employees does not constitute two optional forms &payment. In other words, the same lump-sum option using the same actuarial assumptions on the same terms does not constitute two optional forms of benefit even though there are two different benefit formulas. An example might be a plan that covers salaried employees and hourly employees with two different benefit formulas. The lump-sum option is available on the same terms using the same assumptions and is considered one optional form of benefit.

The regulations specify a partial list of what ancillary benefits might be. There is no hard definition of ancillary benefits. According to the regulations, ancillary benefits include Social Security supplements other than qualified, disability benefits not in excess of a qualified disability benefit (§41 l[a] [9]), ancillary life insurance, health insurance paid out of

374


a plan, death benefits under a DC plan, preretirement death benefits under a DB plan, such as a lump sum, shut-down benefits that are not protected, and finally, all other similar benefits. So when you are reviewing your plans, and if you think you have an ancillary benefit, you probably do even though it's not on the list.

Other rights or features is what I call the IRS catchall. Generally, other rights or features are any fight or feature applicable to employees under the plan. Important exceptions are items already determined to be an optional form of benefit or an ancillary benefit. Also, if the provision is one of the terms that you have taken into account to determine whether you have different optional forms, it is an exception. For example, as you recall from a little earlier, differences in benefit formulas or allocation formulas are considered exceptions. Finally, features that have no meaningful value to employees are exceptions. The IRS gives an example of administrative detail.

One observation of this particular area is that when you do review your plan, you will have to identify every single feature, especially those that are different for groups of employees. That's critical. If you're treating a group of employees differently within the same plan, you had better review the optional forms. Are the ancillary benefits offered in the same manner, or do we have other rights or features that we're offering to one group of employees that we're not offering to another?

The IRS lists some examples of other rights and features. One is the fight to a particular form of investment. An example might be a DC plan that covers two different groups of employees. You offer one family of mutual funds to one group of employees and another family of mutual funds to another group of employees. That might be considered the right or feature by the IRS, and you might have to test it even though the same kind of class of funds are offered.

Another example is the right to make each rate of elective contributions in a 401(k) plan or plans that have different maximum levels of deferral. These could be fights or features that could be considered different.

Other examples include the right to make after-tax employee contributions to DB plans, the right to make alter-tax employee contributions to a profit-sharing plan, the fight to each rate of allocation in a matching plan, such as different matching formulas for different divisions within the same plan, and the right to purchase additional ancillary benefits (life insurance comes to mind for me), and finally, the right, interestingly enough, and one that I 'd probably miss, to make a rollover. A rollover option that is available to one group of employees and not available to another group of employees could result in testing.

In summary, we have DB rights and features as different optional forms of benefit based on relevant terms. Benefits also include ancillary benefits. The regulation has a list of examples and has stipulated that anything that looks similar to an ancillary benefit is one. And then there is the IRS catchall of all other rights and features that are available to any employee under the plan. When you review retirement plans, you have to look for all these different forms and benefits, especially the ones that are going to be offered to a particular or specific group of employees.

375

RECORD, VOLUME 21

So now that we've isolated these BRFs, we're going to proceed to test these provisions. There is a two-pronged test: current availability and effective availability. Current availability is one that we actually can determine.

CURRENT AVAILABILITY The general rule for current availability is a group of employees to whom a BRF is available. It must satisfy §410(b) without regard to the average benefits percentage test. Briefly, what that means is you either have to satisfy the ratio percentage test (the 70% test), or you have to satisfy the reasonableness and nondiscriminatory classification test of §410(b). That means that the classification of employees is deemed to be reasonable by the IRS. An example that the IRS might consider to be reasonable may be hourly-versus- salary classifications. Any bona fide business reason may be a reasonable classification of employees.

The nondiscriminatory nature of that classification is again determined by using the ratio percentage test results. As you recall, under §410(b) there are safe harbor and unsafe harbor percentages. If the ratio percentage is above that safe harbor percentage, the classification is deemed to be nondiscriminatory. If the ratio percentage is between the unsafe harbor percentage and the safe harbor percentage, the classification may be considered nondiscriminatory based on facts and circumstances as reviewed by the IRS. The end result is that if the ratio percentage test is in excess of 20%, the classification might be OK. If the ratio percentage is under 20%, the provision is toast. Remedial action must be taken. I have another comment. If you're using snapshot-day testing, those thresholds are higher. There are some safe harbors adjustments in the substantiation guidelines: 5% for defined- benefit plans and 10% for defined-contribution plans. So be careful of that.

There are vitally important disregarded conditions for purposes of testing current availability. There are certain age and service conditions. This is important when reviewing an early-retirement window. The two IRS examples demonstrate this. An early-retirement provision, such as age 55 and ten years of service, can be disregarded for purposes of testing current availability of that right. By the same token, ifa plan has a time-limited early-retirement provision, let's say, unreduced benefits at 55 with 30 years of service for eligible participants who terminate within six months, you will have to test the window, but you will be able to project the age and service to the end of the window period.

Whether or not you're vested is a disregarded condition for purposes of current availability. The occurrence ofdeath, disability or termination may be disregarded. Youdon' t have to be dead to be considered eligible for a certain ancillary benefit for purposes of testing current availability. You may disregard the fact that those life events have not occurred yet in testing current availability.

You can disregard the fact that filing for a hardship withdrawal or a default on a loan has yet to occur when testing those provisions. The action of applying for or electing an optional form can be disregarded when you're testing current availability.

Regarding family status, the fact that some participants do not have children may apparently be disregarded. An example might be when testing a childrens' death benefit in the plan. Waiver of the Age Discrimination in Employment Act of 1967 (ADEA), another

376


legal right, can be a disregarded condition. Absence of service, such as the requirement of a particular amount of service for a particular form of benefit, can also be disregarded.

Other Items There is a multiemployer exception for an ancillary benefit or other right or feature. It is not an optional form, but an ancillary benefit, other fight or feature. A plan can have a recency test for a particular form of benefit. The IRIS example is a disability benefit to a certain class of employees with a minimum number of hours in the last two plan years.

The implicit condition that the lump-sum value of a benefit be in excess of $3,500 can be ignored when flushing benefits out under $3,500. Some plans may cash out lump sums up to a certain threshold, such as $10,000. The fact that you have a benefit value in excess of $10,000 can be disregarded in testing current availability. Conditions on plan loans such as a minimum balance can also be disregarded in testing current availability.

An important tool for you as a practitioner in the current availability section of the regulations is prospective elimination ofa BRF. Elimination has occurred if the amount or the value of the BRF depends solely on the employee's accrued benefit as of the elimination date. The elimination date is the later of the effective date or the adoption date of the amendment. For DC plans, the accrued benefit apparently can include gains and losses that are assigned to it after the elimination date, unless it's on plan loans or some other nonprotected benefit under a DC plan. The fresh-start compensation adjustment will not cause failure. In other words, you still technically have a frozen accrued benefit, even though you're adjusting the accrued benefit after the elimination date by the fresh-start rule.

The Special Testing Rule The BRF has been eliminated for benefits earned after the elimination date. The BRF with respect to the accrued benefits is currently available as of the elimination date. Then the BRF is deemed to be currently available for all subsequent periods. If we're ahead of the game and we know that we have a problem coming down the pike, we can probably use this tool. But if we're reviewing an ongoing BRF that isn't currently available, we are behind the eight ball, and this tool may not work for us. Also, the BRF cannot be changed after you've had the eliminating amendment or you can't use this tool.

In summary, for current availability the group must satisfy Section 410(b) without regard to the average benefits percentage test. The lowest threshold is the unsafe harbor percent- ageunder §1.410(b)(4)(c) and that's usually 20%. Iftheratio percentage for theclassifica- tion is under 20%, you're toast. You will have to do something. If the percentage is between the unsafe harbor percentage and the safe harbor percentage, you probably have to go to the IRS and demonstrate a nondiscriminatory classification of employees. Some employers may not want to do that. If the percentage is above the safe harbor threshold, the classification is deemed nondiscriminatory, and you'll satisfy 410(b) without regard to the average benefits percentage test, assuming you have a classification of employees that is reasonable.

There are several disregarded conditions; the most important is that you can disregard age and service conditions. Again, this is important for testing early-retirement windows. There is a special rule in this section for prospective elimination. It can be useful, but it

377

RECORD, VOLUME 21

does have a couple pitfalls. The feature must be currently available at the elimination date, and the BRF may not be changed after that date or you can't use this special tool.

EFFECTIVE AVAILABILITY This may be the area where a plan can get nicked on audit. The LRS will not provide a favorable determination on effective availability under Revenue Procedure 93-39. As practitioners, however, we have to look at it for our clients. There is not an empirical test of effective availability, but the regulations stipulate that the determination is based on all relevant facts and circumstances. Generally, a BRF must not substantially favor the highly compensated.

The regulation gives examples of what would be noncompliance situations. The first example is the addition of subsidized early retirement ofunreduced benefits at age 55 aider 30 years of service. The two owners are immediately eligible. Two of nine of the nonhighly-compensated employees can potentially satisfy the requirements. The plan satisfies current availability if you go through the ratios. However, because the highly-compensated owners are actually eligible for this, that could be construed as being in substantial favor of the highly compensated and the plan may not pass the effective availability test.

The next example is an early-retirement window. The plan sponsor adopts an early- retirement window as of December 1 for individuals who terminate by December 31 and doesn't communicate the availability of the window except to a handful of highly compensated. Only the highly compensated employees take advantage of the window. The regulation concludes that this is in substantial favor to the highly compensated and fails the effective availability test.

The last example is where the plan is amended to allow a lump-sum payment to an employee with a certain kind of disability and the only employee who has the disability is the highly compensated owner. Well, the regulation considers that substantial favoritism to the highly compensated and is deemed not to be effectively available.

Briefly, effective availability requires that the BRF must not substantially favor the highly compensated. The regulations provide a couple examples that are fairly obvious in terms of what might be considered to be in substantial favor to the highly compensated. They won' t make a favorable ruling under Revenue Procedure 93-39 in the determination letter process.

SPECIAL RULES There are some special rules that are actually fairly helpful. The first one has to do with mergers and acquisitions. The special testing rule provides that a BRF is currently available and effectively available if the BRF satisfies the current and effective availability standard on the latest date in which an employee of the acquiring group could be hired and that BRF is offered on the same terms as before. It 's different than prospective elimination. You can change the BRF if you're going to replace it with something that's inherently greater in value. We've run across this in the healthcare industry where institutions and sponsored plans are very often merging.

This rule is applied to a BRF accruing under the plan of the current employer and not the prior employer, unless the prior employer's plan is merged into the current employer's

378


plan. You will be determining ratio percentages on a postmerger basis. The example in the regulations is the following: Y acquires X and X maintains a plan with a lump-sum option. Employer Y merges that plan into its plan, which does not provide the lump sum. How- ever, Y continues to provide the lump-sum option to all employees who were acquired on the same terms as before. Therefore, that option is considered currently and effectively available for the plan year of the transaction and all subsequent plan years. So the good feature about this rule is that we can test it once and we're done for all future plan years.

Frozen Participants A frozen participant is a nonexcludable employee with an accrued benefit in the plan who is not currently benefiting. An example might be a group of employees transferred to another member &the controlled group that isn't a sponsor of the plan.

The special-rule requirement is that each BRF that's available to any frozen participant under the plan is separately subject to the current and effective availability requirements.

The regulations provide four rules that you can apply to frozen participants to satisfy the standard: 1. The BRF is currently and effectively available if it were not available to any

employee currently benefiting under the plan. 2. The BRF would be one that is currently and effectively available if all frozen

participants were treated as employees currently benefiting under the plan and the BRF is currently and effectively available.

3. No change in the availability of the BRF may have been made that is first effective in the current plan year, with respect to frozen participants. So you can't change that BRF.

4. Any change that is effective in the current plan year must be made in a nondiscriminatory manner.

Early-Retirement Windows All windows are potential BRFs. A window is disregarded for purposes of §401(a)(4)-4 with respect to an employee for all plan years other than the first plan year. That actually can be helpful if you open up a window for more than one year. I ra significant number of nonhighly compensated go out in that frst year, and you were going to do the test the second year, you could fail miserably, depending on what the termination patterns were.

The fact that different benefit formulas are exempt from being characterized as a different optional form of benefit is going to become vitally important. It means that many windows may not have to be tested as a separate optional form or right or feature. You might have to test under the general test for the benefit amount but not for different BRFs.

The regulations actually gave an example to hang our hats on. A client opens up a window; that adds an extra five years of service to the actual service in determining the benefit. The example specifically indicates that the form of benefit available under the window is not a separate optional form from the early-retirement option form that's available to other employees.

Warning! If you open up a nonformula type of provision in the window, such as a lump- sum option, then you may have to test it. The determination ofa nonforrnula addition to

379

RECORD, VOLUME 21

the plan and a formula addition to the plan could result in some windows falling into a grey area. So you must be careful if you can't really tell whether it's a formula feature that's been changed.

The next special rule actually can be useful, and I 've seen it in practical applications. You can aggregate BRFs together for testing. For example, ifa BRF does not satisfy the requirements of this section, we might be able to aggregate that particular BRF together with another BRF in order to pass. There are two provisions. One of the two BRFs must be inherently greater in value than the other. It has to be impossible for an employee to receive at any time a lower benefit or a less valuable right. Second, the more valuable BRF must satisfy the nondiscriminatory availability. Interestingly enough, you can apply it more than once. You can have sort of a layered aggregation of BRFs.

Example: you have a plan that provides lump sums. One division or one group of employees is using a 5% factor. Another group or division of employees is using a 7% factor. You can prove that the lump sums on the 5% factor are always greater than lump sums on the 7% basis. Let's just assume that the 5% factor lump sum satisfied the requirements of this section. Then you can aggregate the 7% lump-sum optional form with the 5% and test it as one BRF.

If you threw into the mix a 20-year service requirement on the 5% lump-sum factor, it's not so clear that you have a benefit of greater value that could always be available to the participant. So that's when you fail one of those provisions. A participant has to always be able to get the greater form or the greater value of benefit.

Other Special Rules If you're permissively aggregating plans under 410(b), the aggregated plans satisfy this section with regard to the nonsubsidized qualified joint survivor, the spousal death benefit. The subsidies may be determined by using any reasonable actuarial assumption. Spousal death benefits in DC plans are deemed to be nonsubsidized.

Employee stock ownership plans (ESOPs) can have an option that an employee who is at least 55 and has ten years of participation can diversify his or her holdings. There's also a restriction to certain individuals under §409.

We had a client who sponsored a 401(k) plan that was entirely invested in employer stock. The plan had an option that once you reached age 55 and had ten years of participation, you could diversify your holdings. We wanted to use this exception, but we could not because it wasn't an ESOP. The client ended up opening up this right to participants with five years of service. Practically what happened was that everyone in the age-55-and-ten age and service classification diversified their holdings, and no one else did. So we got the same effect that we wanted, but we had to tinker with the plan's provisions because we couldn't apply this special rule.

Unpredictable Contingent Events If you have a BRF that' s contingent on an unpredictable contingent event, you test that provision or that right or feature assuming that the event has occurred. And you're allowed to disregard age and service conditions to the extent allowable as we discussed earlier.

380


In summary, for mergers and acquisitions some special rules will give us prospective compliance as long as we meet certain conditions on the date in which an employee could have been considered in the acquired group. Frozen participants are nonexcludable employees with accrued benefits under the plan who are not currently benefiting. There are four special rules. If the plan can get through one of them, it will comply.

Regarding early-retirement windows, if the window just has a formula change, the plan may be exempt from testing under §401(a)(4)-4. The plan may not be exempt from benefits-amounts testing or plan amendment compliance, but it may be exempt from BRF testing. You must be careful, however, if the plan opens up a window with a nonformula type of right or feature or form that would require testing as a BRF. You can permissively aggregate BRFs as long as one of the BRFs is inherently greater in value and the BRF that is greater in value satisfies the nondiscriminatory requirement of 401 (a)(4)-4.

Other Special Rules The qualified joint survivor annuity (QJSA) and spousal death benefits, if they're nonsubsidized, will have deemed compliance for permissively aggregated plans. There are special ESOP rules for diversification rights and for certain restrictions of ownership. And you're able to test BRFs that are available on unpredictable contingent events. You can assume that the event has occurred and you're allowed to disregard age and service requirements to the extent allowed in testing those particular BRFs.

OTHER ISSUES THAT ARE IN THE REGULATIONS The issue of former participants sneaks up on us. Any changes that are made to BRFs that are available to former participants will probably comply under §401(a)(4)-4 as long as the change is made uniformly with respect to highly-compensated employees and nonhighly- compensated employees and it doesn't discriminate in favor of the former highly-compensated employees. So when you're making those changes to your terminated participants, you do have to pay a little bit of attention to what you're doing there with regard to the former participants.

Qualified Separate Line of Business Gateway You never really think this applies to BRFs, but in fact it does. If you're thinking about using the separate line of business regulations, you should be careful about your BRF. Each BR.F must satisfy the gateway of §410(b)(5)(B), which is our safe and unsafe harbors on an employerwide basis.

EXAMPLES We can go through those examples and then I'll open it up to some questions and some discussion. As we all know, Revenue Procedure 93-39 allows us to answer a series of questions on demonstration 3 for the IRS, so that it can rule on the current availability of certain BRFs. It will not review effective availability at this time.

I think my first example deserves some discussion at least as to what might happen in the future (Table 1). This DB plan offers a lump-sum option to employees who were hired prior to January 1, 1982. It wasn't prospectively eliminated. It's a benefit option that is still available on benefits earned through. Participants had to be hired and enter the plan prior to 1982. The plan eliminated this option to those who were hired after January 1, 1982.

381

RECORD, VOLUME 21

TABLE 1 REV. PROC. 93-39 DEMO. 3 EXAMPLE ONE

RATIO PERCENTAGE TEST (1.410(B)-9)

Employer Subsidiary

! Nonexcludable HCEs 15 100 Nonexcludable NHCEs 133 2,600 Covered HCEs Covered NHCEs HCE Ratio NHCE Ratio

~Ratio Percentage Test Safe Harbor Threshold Unsafe Harbor Threshold

8 0 47 0

0.53 0.35

66.26% 28.25% 20.00%

Total

115 2,733

8 47

0.69 0.17

24.72% 23.75% 20.00%

For the 1994 plan year, the plan had a ratio percentage test of 66%, well above the safe harbor threshold of somewhere in the 30% range. We made an argument that the group of employees make up a reasonable classification. The concentration percentage is 89%, so the safe harbor percentage was 2825%. Because 66% is in excess of 28.25%, we automatically satisfy the nondiscrimination requirement under §410(b)-4(c)(4), and there we made the argument that it satisfies §401 (a)(4)-4(b). Some day that 66% will decrease, depending on the termination patterns of the employees who are still eligible.

Then what will the plan sponsor do? The plan sponsor will probably eliminate it prospectively. That'll be the easiest thing to do when the ratio percentage is down near the safe harbor percentage. Depending on who's left, that could be a sticky issue if highly compensated individuals are left and want the option. But you can see that prospective elimination will probably work here.

Example 2: let's say a plan sponsor calls and informs you that there is a previously unknown controlled group situation with a substantial subsidiary. You couldn't believe it. You didn't think the plan sponsor was in a controlled group of employers. Well, perhaps qualified separate line of business designation will work and the plan will be OK. And yet we have this lump-sum option that we're going to have to test on an employerwide basis.

You can see in Table 2 that when we didn't think the plan sponsor was in a control group, the plan had a 66% ratio percentage. That's well in excess for the safe harbor, so for that particular plan year, the plan looked good. With the addition of the subsidiary we have a ratio percentage of 24.72%. The safe harbor threshold based on this concentration percentage is 23.75%, and the plan barely passes. What if we were using snapshot-day testing? That 23.75% safe harbor percentage would be (this is a DB plan) bumped up 5%. All of a sudden the ratio percentage of 24.72% is not above the safe harbor threshold. I might squeak by ifI use quarterly testing or something such as that, or ifI can manage to get annual data on all members of the controlled group, which we all know is the easy thing to do.

382


TABLE 2 REG. § 1.410(B)-9 RATIO PERCENTAGES AS OF 12/31/94

Nonexcludable HCEs Nonexcludable NHCEs Covered HCEs Covered NHCEs HCE Ratio NHCE Ratio Ratio Percentage Test

RPT Threshold DB Snapshot

Safe Harbor Threshold DB Snapshot

Unsafe Harbor Threshold DB Snapshot

Employer

15 133

8 47

0.53 0.35

66.26%

70.00% 73.50%

28.25% 29.66%

20.00% 21.00%

Subsidiary

100 2,600

0 0

Controlled Group

115 2,733

8 47

0.6957 0.1720

24.72%

70.00 % 73.50%

23.75% 24.94%

20.00% 21.00%

You can see that we're right on the threshold of not passing. In fact, it wouldn't take many bodies to put us under the unsafe harbor threshold. In fact, I calculated the DB snapshot safe harbor threshold of 24.94%. That's close to 24.72%. Unfortunately, we must carry out the four decimals here. Even though we're within two tenths of a percent, it 's still not going to pass if we're using snapshot-day testing. You can see that if you don't know if you're in a controlled group, it can come up and bite you.

Example Three: a couple clients were going through quite a few mergers and acquisitions. This particular client actually merged two plans into its salaried plan, and it was straightfor- ward. It went ahead and froze the benefits on the elimination date, and each of them passed the ratio Percentage test on the elimination date. It is not going to make any changes; we'll probably be able to say we passed with respect to those benefits for prospective periods. The ratio percentages take into account all employees of the employer.

Example Four: we had one plan that offered a Medicare supplement and the plan eliminated that supplement prospectively. The provision's ratio percentage barely passed the safe harbor percentage. This is very similar to the situation we had in which we looked at the qualified separate line of business in the first example (Table 3). This was a smaller entity in this particular controlled group, and you can see our ratio percentage was fairly low at 32.78%. The employer had a concentration percentage of 96%, so the safe harbor percentage was 23%. So we satisfied §l.410(b)-4(b)(3),

Example Five: this is one instance where we did use the aggregation of BRF. We have a plan that has that two definitions ofnorrnal retirement. Normal retirement is the first day of the month after your 65th birthday or your 62nd birthday, if you have 25 years of service, So what do we do here? We disregarded the age and service conditions in testing the current availability, and we aggregated the 62 normal retirement date (NRD) feature

383

RECORD, VOLUME 21

with the 65 NRD (Table 4). The NRD at age 62 is of inherently greater value. Therefore, we satisfied the ratio percentage test. We had a 97% ratio percentage test, and we satisfied § 1.401 (a)(4)-4(d)(4)(B).

TABLE 3 REG. § 1.410(B)-9 RATIO PERCENTAGES AS OF 12/31/92


496 13,924

18 165

0.0363 0.0119

32.78%

TABLE 4 REG. § 140(B)-9 RATIO PERCENTAGE TEST AS OF 12/31/92


8 42

8 41

1.0000 0.9762

97.62%

MR. DENNIS M. MORRIS: My question may not be to the point of some of the material you're covering, but I 'm asking this because I don't practice in this area. When the IRS set up a safe harbor percentage, I 'm curious as to the logic of why it would develop a percentage that would be so much lower than the 70%. Do you have any background in how it went about setting up the formulas to develop these safe harbor percentages?

MR. STURDIVAN: No.

FROM THE FLOOR: You gave the example in which these individuals had the lump-sum option. The plan could actually continue to qualify so long as 19 nonhighly compensated employees were covered. Based on the logic of the 70% test, it just didn't seem to compute.

MR. STURDIVAN: I don't know what the theory was behind setting the threshold so much lower, but it does give you some wiggle room.

FROM THE FLOOR: I 've a question about your last example on the normal retirement date in which you have different normal retirement ages depending on service. Were you only testing this because it was a certain group? I 'm somewhat confused. Is the entire plan population eligible for this provision?

MR. STURDIVAN: The plan has the language and it does not specify a separate group for eligibility. But it was a different normal retirement age so we had to test it.

384

Documents

Benefits, Rights, and Features · benefits available under the plan. Rights and features are all other rights and features available to any employee under the plan. So it's not the